Enhancing Change Detection and Model Comprehension in Parametric Design Systems
Change detection, control and comprehension are important tasks in parametric design for agile analysis and informed decision-making. Increasing complexity of parametric systems interfaces and design models, combined with human visual perception limitations can negatively influence designers' performance.
In this thesis, I investigate change detection and human visual perceptual mechanisms in the context of parametric design. I propose a set of debugging-like interaction techniques on dataflow graph interfaces to assist with change detection and model comprehension by enabling designers to identify data flow effects, parametric dependencies, and changes.
The work presents a series of sketches, interactive demonstrations, as well as a high-fidelity interactive prototype to evaluate the effectiveness of the techniques. The results of the tasks-based talk-aloud user studies support the proposed design and reveal valuable recommendations for further improvements.
The thesis provides generalizable knowledge for change detection and control that can be used in other systems showing similarities to parametric system interfaces.